Automated water quality monitoring and regression-based forecasting system for aquaculture

Water quality in fish tanks is essential to reduce mortality. Many factors affect the water quality, such as pH, dissolved oxygen, and temperature tanks. Existing work has presented monitoring systems for aquaculture, which are useful automatic notify any incidence of decline quality. It enables farms make interventions However, advanced through forecasting necessary ensure consistent optimum This paper presents a web-based system aquaculture. First, model based on long short-term memory designed developed. The evaluated fine-tuned using existing public dataset. Second, prototype An Arduino Raspberry Pi data acquisition tool built. A application then developed present forecasting. notification module included send an alert message farmers when necessary. tested at hatchery Universiti Malaysia Sabah. findings show that proposed provides better management farms.